Troubleshooting
Problem
How do I install the open-mpi wrapper for LAVA?
Resolving The Problem
How do I install the open-mpi wrapper for LAVA?
Please find the procedure below for this installation of the attached wrapper:
1- create a lava.custom directory under /share/apps :
sudo mkdir /share/apps/lava.custom
2- create / copy the openmpi-mpirun script
sudo cp ./openmpi-mpirun /share/apps/lava.custom/
3- publish it on the frontend :
sudo ln -s /share/apps/lava.custom/openmpi-mpirun /opt/lava/6.1/linux2.6-glibc2.3-ia32e/bin/openmpi-mpirun
4- push it on the nodes :
sudo cluster-fork "ln -s /share/apps/lava.custom/openmpi-mpirun /opt/lava/6.1/linux2.6-glibc2.3-ia32e/bin/openmpi-mpirun"
5- check everything was well propagated :
cluster-fork "which openmpi-mpirun"
6- Check the wrapper with the latest version of openmpi (1.2) :
module load hpc/ompi12-gnu
mpicc -o cpi_test cpi.c
bsub -o%J.out -n 8 openmpi-mpirun -np 8 ./cpi_test
mpicc -o cpi_test cpi.c
bsub -o%J.out -n 8 openmpi-mpirun -np 8 ./cpi_test
-----------------JOB OUTPUT-----------------
[toor3@frontend run]$ cat 2496.out
Sender: LSF System>
Subject: Job 2496: Done
[toor3@frontend run]$ cat 2496.out
Sender: LSF System
Subject: Job 2496:
Job was submitted from host by user .
Job was executed on host(s) <2*compute-1-6>, in queue, as user .
<2*compute-1-15>
<2*compute-1-9>
<2*compute-1-1>
was used as the home directory.
was used as the working directory.
Started at Fri Apr 13 07:50:43 2007
Results reported at Fri Apr 13 07:50:46 2007
Job was executed on host(s) <2*compute-1-6>, in queue
<2*compute-1-15>
<2*compute-1-9>
<2*compute-1-1>
was used as the home directory.
was used as the working directory.
Started at Fri Apr 13 07:50:43 2007
Results reported at Fri Apr 13 07:50:46 2007
Your job looked like:
------------------------------------------------------------
# LSBATCH: User input
openmpi-mpirun -np 8 ./cpi_test
------------------------------------------------------------
# LSBATCH: User input
openmpi-mpirun -np 8 ./cpi_test
------------------------------------------------------------
Successfully completed.
Resource usage summary:
CPU time : 0.15 sec.
Max Memory : 1 MB
Max Swap : 13 MB
Max Memory : 1 MB
Max Swap : 13 MB
Max Processes : 1
Max Threads : 1
Max Threads : 1
The output (if any) follows:
Process 0 on compute-1-6.local
Process 2 on compute-1-15.local
Process 4 on compute-1-9.local
Process 6 on compute-1-1.local
Process 3 on compute-1-15.local
Process 5 on compute-1-9.local
Process 7 on compute-1-1.local
Process 1 on compute-1-6.local
pi is approximately 3.1416009869231241, Error is 0.0000083333333309
wall clock time = 0.012229
Process 2 on compute-1-15.local
Process 4 on compute-1-9.local
Process 6 on compute-1-1.local
Process 3 on compute-1-15.local
Process 5 on compute-1-9.local
Process 7 on compute-1-1.local
Process 1 on compute-1-6.local
pi is approximately 3.1416009869231241, Error is 0.0000083333333309
wall clock time = 0.012229
-----------------JOB OUTPUT-----------------
[{"Product":{"code":"SSZUCA","label":"IBM Spectrum Cluster Foundation"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"--","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"4.1.1","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}},{"Product":{"code":"SSZUCA","label":"IBM Spectrum Cluster Foundation"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":null,"Platform":[{"code":"","label":""}],"Version":"","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]
Was this topic helpful?
Document Information
Modified date:
11 September 2018
UID
isg3T1014105